NOTES
258
VOL. 28
TABLE I. 5-(Substituted) Amino-1,2,3-thiadiaeoles
N-CH Yield,
R=
M.P., 0C.U
Crystalline' form
%d
7-NFormula
Calod.
S--7
7 -
Found
Calcd.
Found
CeHs 42.0 Grey needles 180 GH?NsS 23.71 23.80 18.09 17.80 C&6CHzb 7.3 White needles 93-95 CsHgNsS 22.19 22.21 16.75 16.52 13.80 20.0 Tan powder 23.80 14.42 4OzNCeH4 C~HBN~SO~ 25.18 206-209 4-CHaOCeH4 10.0 Colorless flaked 155-157 CSH~NSSO 20.30 20.40 15.48 15.10 10.0 Tan powder 173-175 C8H6NaSClh 19.85 4-ClCBH4 15.17 15.71 19.60 13.0 Brown flakes 187-189 C8HoN3SBri 16.40 16.15 12.50 12.77 4-BrCsH4 35.5 Yellow needles CgHgNaS 22.19 22.40 16.75 16.38 4CHaCeHd 172-174 23.0 Green powder 14.53 14.00 4-(CHs)2NCeHrlC 24.60 CIOH12N,S 25.42 168-170 13.55 14.26 20.0 Yellow needles 161-162 C12HgNaS 18.50 18.35 CioH?' 37.0 Tan scales 165-166 CsH8N3SBr' 16.80 12.50 12.85 3-BrCeH4 16.40 a All are new compounds with exception of R = CBHS (lit.,6 m.p. 179-180°). Product did not precipitate; ether was removed and Product did not precipitate; ether was removed and product precipitated by benzene. residue recrystallized. Represent first With decomposition except R = C6H6CH2. C1: calcd., 16.78; found. crop of crystals. e From ethanol. f From chloroform. Br: calcd., 31.17; found, 31.40. IC a-Naphthyl. 16.90. Br: calcd., 31.17; found, 31.15.
'
TABLE I1 ULTRAVIOLET ABSORPTION SPECTRA O F 5-(SUBSTITUTED) AMINO-1,2,3-THIADIAZOLES
N/~'C-NH--R~ I1
II I1
N-C-R;! Log
RZ H H
hmax
tmax
240 246
4.54 4.21
H
258
3.94
H H
246 232
3.90 4.47
C6Hs
236
4.05
H H
240 230 240 252 260
3.96 4.06 4.03 3.93 4.34
H H
Log Xmax
tpuzx
LOP Xmax
266 4.48
236 240 242 254 274 280
4.27 4.28 3.89 3.84 3.71 3.66
296 314 324 310 330 324 334
~ l m X
4.53 4.34 4.36 4 26 3.97 4.21 4.21
316 3.49 318 4.05 320 3.77 320 4.24 324 4.17
TABLE I11 MAJORCHARACTERISTIC INFRARED FREQUENCIES OR 5-( SUBSTITUTED) AMINO-1,2,3-THIADIAZOLES Cm.-15
Assignment
3220 m Bonded N-H stretching 1650-1590 v N-H deformation 1560-1475 V' Ring stretching 1350-1280 v d C-N stretching 1265-1200 v C-H in-plane deformation 1190-1175 v d C-H in-plane deformation 1150-950 vd Ring breathing 910-890 w Ring breathing 705-670 w C-H out-of-plane deformation a Intensity: m, medium; w, weak; v, variable. One or more bands. Two bands. At least one band.
Some Reactions of Fluorinated Cyclobutenes with Grignard Reagents J. D. PARK AND R. FONTANELLI' Department of Chemistry, University of Colorado, Boulder, Colorado Received April 20, 1962
I n an attempt to find a simple method for introducing alkyl and aryl substituents into fluorocyclobutenes, our
attention was turned to the reaction of Grignard reagents with these compounds. This had not been carried out previously. The reaction of Grignard reagents with fluoroolefins, such as CF2=CC12, CFZ=CFCl, and CF&l-CF=CF*, has already been described by Tarrant, et aZ.* These workers found that apparently addition first occurs across the double bond and the resulting adduct loses MgXz to give a new, longer chain fluoroolefin. The reaction goes with poor yield (10-2070) using aliphatic Grignard reagents and with better yields (30-70y0)using aromatic Grignard reagents. The preparation of some alkyl derivatives (mono- and dimethyl, mono- and dibutyl, and diphenyl) of perfluorocyclobutene, has been previously described by D i ~ o nusing , ~ the reaction with alkyllithium. This reaction, however, gives in poor yields (20-40%) only the diphenyl derivative with phenyllithium and L mixture of mono- and dialkyl derivatives (the latter predominating) with alkyllithium. I n this study, when the perfluorocyclobutene was treated with excess alkylmagnesium bromide under mild conditions, the monoalkyl derivatives in high yields (75-85%) (methyl excepted) had been obtained. Under stronger conditions, the monoalkyl derivatives with excess Grignard reagent gave comparable yields of the dialkyl derivatives. However, the reaction with phenylmagnesium bromide gave both the mono- and diphenyl derivatives (ratio 1 : 1) in a total yield of 80%. Some reactions with vinyl- and perfluoroalkylmagnesium bromide have been attempted, but only high boiling polymeric materials were isolated. When 1,2-dichloro-3,3,4,4-tetrafluorocyclobutene mas treated with alkyl Grignards, the substitution of one vinylic chlorine took place quite readily. The substitution of the second vinylic chlorine, on the contrary, does not easily take place, even after refluxing the monoalkyl derivative for twenty-four hours in ether. This reaction has been carried out in a sealed Pyrex tube in ether under autogenous pressure, and only above 100' a reaction took place. The reaction products isolated were: starting material ('25yo\,1.2-diethyl-3,3,4,4-tetrafluorocyclobutene (16%) (this product was identical (1) Montecatini Industrial Reeearch Fellow (1961-1962). Work done a t the Univereity of Colorado. (2) P. Tarrant a n d D. A . Warner, J. Am. Chem. Soc., 76, 1624 (1954). (3) S. Dixon, J . O w . Cham., 21, 400 (1956).
NOTES
SANUARY, 1963
259
TABLE I PHYSICAL PROPERTIES OF THE CYCLIC ETHERS Yield,
CFt-C-E
t
CF2-C-F CF2-C-E
t
II
I
n35D
dz54
----
Calcd.--
7
B.p., 'C./mm.
C7u
Conipounds
C
H
F
C1
C
H
Found----F
C1
75
66-68/630
1.2421
1.3303
41.87
2.93
55.20
41.52
2.78
55.12
75
134-136/630 69-70/60
1.110
1.3735
52.74
5.53
41.72
.53.03
5.78
41.62
83
86-88/630
1.199
1.3421
45.17
3.79
51.04
44.95
4.00
51.28
CF2- *-C3H7 CF2-C-CcHb
76
77/25
1.060
1.3908
57.13
6.71
36.15
57.28
6.59
36.01
CF2-C-F CFz-C-CH3
80
67-68/15
1.354
1.4606
54.56
2.27
43.15
54.85
2.47
43.24
80
77-78/630
1.337
1.3602
36.03
1.73
43.54
20.32
36.00
1.92
43.39
20.61
7.5
98/630
1.292
1.3723
38.21
2.67
40.30
18.81
38.46
2.65
40.21
19.02
CF2-k-Cl CFz-C-CsHj
78
118/630
1.236
1.3795
41.50
3.49
37.51
17.50
41.53
3.70
37.71
17.41
CFr-C-C1
65
78-80 /5
1.371
1.5012
50.77
2.13
32.12
14.99
51.00
2.50
32.31
14.49
CFZI & --Et, CFz-C+-CzH7
CFS-C-GH7
I
&
I
11
III:
CF2- -C1 CF2-C-C2H '
I1
I
li
I
!I
i
CFz-C-Cl CF~-C+-CZH?
to that already obtained from the reaction of l-ethyl2,3,3,4,4-pentafluorocyclobutene with ethylmagnesium bromide) and a difficultly separable mixture of five high boiling products that were partially resolved by vapor phase chromatography, but not identified. These may be a mixture of the triethyl derivatives formed from the substitution of the allylic fluorines. What has been found in this study is in agreement with the base-catalyzed reactions of alcohols with perfluorocyclobutene and 1,2-dichloro-3,3,4,4-tetrafluorocyclobutene.4v6 Proof of Structure.-In the infrared spectra of the mono- and dialkyl derivatives of perfluorocyclobutene, there is evidence that the substitution of the vinylic fluorines took place. Thus, the double bond of perfluorocyclobutene absorbs a t 1790 cm.-l and the monoalkyl derivatives exhibit a shift of the double bond absorption band to 1720-1730 cm.-l. The dialkyl derivatives do not exhibit any absorption (or very weak a t 1705 ern.-') as the tetrasubstituted double bond is perfectly symmetrical. The reaction probably goes through the addition of the reagent (RMgX) to the double bond, and the intermediate product thus formed loses the magnesium halide giving the substituted olefin. CF2-CF
-
CFz-CF( R )
RMgX
hF*-AF(MaX) .
--+ CF2-C-R
1
/I
CF2-C-F
+ MgXF
However, in the reaction of 1,2-dichloro-3,3,4,4-tetrafluorocyclobutene with Grignard reagents, the first step goes in the same way, but the displacement of the second chlorine is much more difficult because of the lesser tendency of chlorine to undergo mesomeric shift. (4) J. D.Park, M. L. Shsrrah, and J. R . Laoher, J. Am. Chem. Soc., 71, 2337 (1949). (6) J. D.Park, C . M. Snow, and J. R . Lecher, ibid., 78, 2342 (1961).
In such a case, the vinylic chlorine atom shows the usual inertness toward displacement. Experimental
l-Methyl-Z,3,3,4,4-pentafluorocyclobutene(I).-In a 500-ml. three-neck flask fitted with a gas inlet tube, a stirrer, and a Dry Ice condenser connected to a bubbler, 250 ml. of 3 M ethereal solution of methylmagnesium bromide (0.75 mole) was introduced and cooled to 0'. A 50-g. sample (0.30 mole) of perfluorocyclobutene was bubbled into the ethereal solution in a period of 1 hr. and further maintained for 2 hr. a t 0" and for another 20 hr. a t room temperature. The reaction mixture was then warmed to gentle reflux of the ether for 4 hr. After cooling again to 0", 150 ml. of 20% hydrochloric acid was dropped in very slowly to decompose the Grignard reagent. The decomposition is very exothermic and a strong evolution of methane occurred. The ethereal solution was separated and the aqueous layer extracted three times with ether. The combined extracts were washed with bicarbonate solution and dried over anhydrous sodium sulfate. Distillation yielded 11 g. (22%) of I; b.p. 44-45"/630 mm.; n 4 1.3225; ~ d4a1.377; molecular refraction: calcd. for C5HaFa, 23.44; found, 22.92. Dixon3 reports a boiling point of 50"/730 mm. for this compound. The preparation of the other derivatives in general followed along similar lines. These reaction products and their properties are tabulated in Table I. Reaction of l-Ethyl-2-chloro-3,3,4,4-tetrafluorocyclobutene with Ethylmagnesium Bromide.-In a high pressure Pyrex tube fitted with a valve and a pressure gage, 100 ml. of a 3 M ethereal solution of ethylmagnesium bromide (0.3 mole) and 18.5 g. (0.1 mole) were introduced. The tube was then warmed to 100-110" for 30 hr. The formation of a grayish precipitate was noticed after a few hours and the pressure had increased to about 120 p.5.i.g. After cooling, the residual pressure was discharged and the content treated with 100 ml. of 20% hydrochloric acid. The ethereal layer was separated and the water layer extracted three times with ether. The combined extracts, washed with bicarbonate solution and dried over sodium sulfate, were distilled. After removal of the ether, 18 g. of a high boiling residue was obtained. Fractionation of the residue under reduced pressure yielded 6 g. of starting material, 4 g. of I11 (b.p. 68-75'/60 mm.) and 7 g. of a fraction distilling a t 70-80"/0.5 mm. which upon vapor phase chromatography was resolved into five products which as yet have not been identified (column: silicon D/C/710, temp. = 200").
NOTES
260
VOL.28
Acknowledgment.-The authors are indebted to Montecatini Industries of Milan, Italy, for making it possible for one of us (L.F.) to have a leave of absence and support to carry out this work at the University of Colorado.
hydroxyquinolines, IIa,b,c, in good yields. 3,4-Dihydroxyquinoline (IIIa) and 3,4-dihydroxyquinaldine (IIIb) were obtained from Dakin oxidations of I I a and I I b with sodium hydroxide and hydrogen peroxide in satisfactory yields. Oxidation of IIIa and IIIb was accomplished with silver oxide and/or chromium trioxide. The products have been assigned the structure A Direct Synthesis of 4-A~anaphthoquinones-1,2~ of 4-azanaphthoquinone-1,2 (IVa) and 3-methyl-4azanaphthoquinone-1,2 (IVb), respectively. Condensation of the azaquinones with o-phenylenediamine gave L. R. MORQAN, JR., AND R. J. SCHUNIOR the corresponding phenazines supporting the initial assignments of the formyl group in IIa and IIb. Department of Pharmacology, Louisiana State University, I n sharp contrast to the foregoing results, the Dakin School of Medicine, New Orleans, Louisiana oxidation of 2-phenyl-3-formyl-4-hydroxyquinoline (IIc) AND J. H. BOYER with sodium hydroxide and hydrogen peroxide to 2phenyl-3,4-dihydroxyquinoline (IIIc) was unsuccessAmerican Chemical Society, Petroleum Research Fund, ful. The major product was the anthranil of phenylWashington, District of Columbia glyoxylic acid (V) whose structure was confirmed by hydrolysis to phenylglyoxylic acid. Apparently in Received J u l y 18, 1962 addition to the Dakin oxidation of the formyl group in IIc a Baeyer-Villiger transformation occurs with The difficulty encountered in the attempt to prepare oxidation of an intermediate peroxide and ring fission. azabenzoquinones2 by the oxidation of hydroxy- and All attempts to stop the reaction at the dihydroxy stage amin~pyridones~ suggested that fusion of a benzene mere unsuccessful. ring to either an 0- or p-azabenzoquinone might increase their stability. The reported synthesis and Everimentale stability of 4-azanaphthoquinone-1,2 (IVa)4 prompted Preparation of the 4-Hydroxyquinolines (Ia-c).-4-Hydroxyan investigation of the chemistry of azanaphthoquiquinaldine,' 2-phenyl-4-hydroxyquinoline,*and 4-hydroxyquinonones. The present paper describes the preparation line9 were prepared according to the literature cited. of two 4-azanaphthoquinones-l,2,IVa and IVb. Preparation of the 3-Formyl-4-hydroxyquinolines(IIa-c) .-
The synthesis of IVa and IVb now reported utilizes the extension of the Reimer-Tiemann reaction to hydroxyquinolines. This approach finds analogy in the observation of Bobranskis that 4-hydroxyquinoline and 4-hydroxyquinaldine are formylated with sodium hydroxide and chloroform. Unfortunately a rigid structure proof of the products was not provided in either case. Formylation of Ia, b, c under Bobranski's conditions proceeded as desired with formation of 3-formyl-4-
3-Formyl-4-hydroxyquinoline6 and 3-formyl-4-hydroxyquinaldine10 were previously prepared. A mixture of 2-phenyl-4-hydroxyquinoline (2.87 g., 0.013 mole), 2 g. of powdered sodium hydroxide, and 2 ml. of chloroform was heated a t 50' for a few minutes and 3 ml. of water added. The slurry was gently refluxed for 6 hr. with 2 ml. of chloroform being added a t 2-hr. intervals. The excess chloroform was removed in vacuo and the resulting slurry filtered. The dried solid was extracted twice with 20-30 ml. of hot water and the washings combined with the original filtrate. Acidification with glacial acetic acid afforded a yellow suspension which precipitated as a yellow sirup that solidified upon standing. Several recrystallizations from ethanol afforded yellow needles of 2-phenyl-3-formyl-4-hydroxyquinoline,m.p. 250-252', 1.2 g. (37%). Anal. Calcd. for CI~HIINOP:C, 77.09; H , 4.45; N, 5.63. Found: C, 77.27; H , 4.52; N, 5.57. The aldehyde formed a 2,4-dinitrophenylhydrazone which recrvstallized from ethvl acetate and ethanol as red needles, m.; 275-277.5' dec. Anal. Calcd. for CzZHlsN6Os: C, 61.54; H , 3.52; N , 16.31. Found: C, 61.27; H , 3.45; N , 16.41. General Procedure for the Dakin Oxidation of 3-Formyl-4hydroxyquinolines (IIa-b).-To a solution of 0.007 mole of the 3-formyl-4-hydroxyquinolines in 7 ml. of 1 N sodium hydroxide, 9.5 g. of 3% hydrogen peroxide was added in one portion and allowed t o stand overnight a t room temperature. A color change from deep orange to yellow was accompanied by an exothermic reaction. Upon cooling t o room temperature the dihydroxyquinolines could be isolated. 3,4-Dihydroxyquinoline recrystallized from 95% ethanol as yellow microcrystals, m.p. 222-227' dec., 18% yield. Anal. Calcd. for CgH,N02: C, 67.07; H, 4.38; N, 8.69. Found: C, 67.13; H, 4.21; N , 8.75. 3,4-Dihydroxyquinaldine recrystallized from 95% ethanol as a pale yellow powder, m.p. 275-281' dec., 49% yield.
(1) Part of this work was carried o u t in the Department of Chemistry, Tulane University, New Orleans, La. (2) In the present study, aaaquinones designates nitrogen as a member of the quinone ring. (3) J. H. Boyer and S. Kruger, J. Am. Chem. Soc., 79, 3552 (1957). (4) M. Passerini, T. Bonciani, and N. Di Gioia. G a m chim. itol., 61, 959 (1931). (5) B. Bobranski, Chcm. Ber., 69, 1113 (1888).
(8) Semimicro analyses by Alfred Bernhardt Microanalytisches Laboratorium, Max Planck Institute, MUlheim (Ruhr), Germany. Melting points are uncorrected. (7) G. A. Reynolds and C. R. Hauser, "Organic Syntheses," Coll. Vol. 111, John Wiley and Sons. Inc.. New York, N . Y., 1955, p. 593. ( 8 ) R. C. Fuson and D. M. Burness, J . Am. Chem. SOC.,61, 2890 (1939). (9) R. G. Gould, Jr., and W. A. Jacobs, ibid., 61, 2890 (1939). (10) M. Conrad and L. Limpach, Chem. Ber., 21, 1965 (1888).
OH
0
Ql$c R IVa. R = H b. R = CHs IIa. R = H, R' = CHO Ia. R = R' = H b. R = CHI, R' = H b. R = CHI, R' = CHO C. R = C&, R' = CHO C. R C&, R' = H IIIa. R = H, R' = OH CH,, R' = OH b. R C.
R
=
Ce&, R' = OH
I
v
